MS2SL: Multimodal Spoken Data-Driven Continuous Sign Language Production
This work addresses communication barriers between sign and non-sign language users by enabling direct sign language production from spoken inputs, representing an incremental advancement in multimodal AI applications.
The paper tackles the problem of generating continuous sign language sequences directly from spoken content like text or speech, proposing a unified framework that uses a sequence diffusion model and joint embedding space to achieve competitive performance on datasets such as How2Sign and PHOENIX14T.
Sign language understanding has made significant strides; however, there is still no viable solution for generating sign sequences directly from entire spoken content, e.g., text or speech. In this paper, we propose a unified framework for continuous sign language production, easing communication between sign and non-sign language users. In particular, a sequence diffusion model, utilizing embeddings extracted from text or speech, is crafted to generate sign predictions step by step. Moreover, by creating a joint embedding space for text, audio, and sign, we bind these modalities and leverage the semantic consistency among them to provide informative feedback for the model training. This embedding-consistency learning strategy minimizes the reliance on sign triplets and ensures continuous model refinement, even with a missing audio modality. Experiments on How2Sign and PHOENIX14T datasets demonstrate that our model achieves competitive performance in sign language production.